Talk:Trading

I just created a cartographer villager (with experimental trades turned on) and promoted him to apprentice. I got two maps listed. This is a change from 1.20 vanilla, but not unexpected.

I wonder if a note should be made in the trade rebalance section to note that a single villager can give two maps at the same level. SirDaddicus (talk) 23:56, 13 November 2023 (UTC)

I went into the editor to add something to note 6, but it wouldn't let me. How does one edit a note? SirDaddicus (talk) 15:13, 15 November 2023 (UTC)

Try editing the source, on the top under the 'help' tab find 'references' or something like that. Each note should have a definition somewhere, but I can't find the definition for the enchant-1.8 note (Note 2). Can someone tell me where that is? I tried testing in the sandbox, or in the help page, but nothing seems to work. I can try to help you edit note 6 though. MinecraftExp123 (talk) 05:04, 14 February 2024 (UTC)

@MinecraftExp123: Note 2 is defined in line 4402. Kationor (talk) 07:53, 14 February 2024 (UTC)

Thank you very much! MinecraftExp123 (talk) 09:17, 14 February 2024 (UTC)

Editing the notes won't work, because one still must update the table. Plus, the table is just wrong. There should be three trades under the apprentice level, similar to the journeyman level. SirDaddicus (talk) 16:19, 6 August 2025 (UTC)

Someone who knows how to edit tables needs to fix the cartographer table. It's wrong. SirDaddicus (talk) 19:23, 8 August 2025 (UTC)

The Armorsmith trade rebalance table is missing entries. It only shows expert- and master-level trades, but all the trades changed. (For example, the first trades are asking for iron ingots and coal.)

I have no idea how to change a table, though. Can someone please take a look? Thanks! SirDaddicus (talk) 18:23, 14 November 2023 (UTC)

I know how to change the table, but I don't know the other trades. Please send me a file with those trades if you have them, and I will edit the table. MinecraftExp123 (talk) 13:45, 6 February 2024 (UTC)

I think I found the trades on the feedback.minecraft.net page, but it doesn't say the amount of experience the new trades give to villagers. Most of the trades have changed, and probably so did the amount of XP. Can someone find a reference or just an in-game testing about that? I see all other tables on this page has the number of XP the trade gives to villagers. Also that IP user is actually me, when I forgot to sign in. MinecraftExp123 (talk) 01:37, 7 February 2024 (UTC)

I've added a link to the Villager Trade Rebalance page, saying that's the main article (using the Main Article template). All the information is there, but I don't want to copy all those tables here as it could cause lots of lag. I hope someone can do this, thanks! (My computer isn't very good when editing large pages) MinecraftExp123 (talk) 14:11, 14 February 2024 (UTC)

Also, I think some of the Toolsmith and Weaponsmith tables are missing information. I don't know how to edit table either, and can someone add the iron block and diamond block to emeralds trade? 221.126.252.146 03:20, 29 January 2024 (UTC)

The cartographer expert level trade lists the item frame as a 100%, however this is not true in practice likely due to the behaviour of the other expert level trade option - any colour banner - which individual offers can be given for different colour banners. I know there is a note that explains that this can be the case, but this should also be reflected in the probability values as they can be misleading otherwise.

Zafe (talk) 03:44, 29 July 2024 (UTC)

Hmm. I observed some strange behavior with the trade rebalance in the same place, and I can't figure out what's up. -~- Nerdyguy2000   talk   edits 14:45, 29 July 2024 (UTC)

👁 Image
 Fixed by splitting the banner trade. Note that it looks kind of ugly now, so we might need to consider a different solution, but it's all I can do for now. --MinecraftExp123^{(talk|contribs)} 02:09, 30 July 2024 (UTC)

@Zafe and Nerdyguy2000: I understand the problem now. The thing is, some of the sheperd trades and maybe some other ones also share the problem, so we need someone to either split all those trades or create a bot to do so. I can't do it right now, because I've got two other things to work on: converting Villager Trade Rebalance to use {{TradeTable}} and {{TradeLine}}, and completing the tables at Damage#History. If you have the time, please start fixing these trades! --MinecraftExp123^{(talk|contribs)} 12:23, 12 August 2024 (UTC)

Note: I meant "Shepherd" when I said "Sheperd". --MinecraftExp123^{(talk|contribs)} 12:35, 13 August 2024 (UTC)

It says on the page that 'it's impossible to get a baby nitwit by breeding two nitwits together". Babies can't appear as nitwits, but they do have a 10% to grow into a nitwit in bedrock. Can I add this?

Another question I have: I keep seeing things like villagers being able to give items with enchantements levels "5-19". I'm assuming this is referring to the amount of bookshleves around an enchantment table, but I can't find any information about which enchantments are whichever levels. What would be an example of a "level 5 enchantment" compared to a "level 19 enchantment"?

SquebbyNICO (talk) 18:31, 7 August 2024 (UTC)

For the first question: that is actually already listed in the Villager article. I'm not sure why there is anything about breeding nitwits on this page at all, as this article is supposed to be about the trading mechanic; I might even consider deleting that line from the article, since it could be considered irrelevant. For the second question, see Enchanting/Levels. It has to do with the technical way that the game determines which enchantments to offer. On that page, each enchantment has a range of numbers associated with it. If that range of numbers contains a number between 5 and 19, then it is a possible enchantment that could be offered on a traded item. So for example, an armorer villager could offer armor enchanted with Protection I or II, but not III or IV. Rampage455 (talk) 19:04, 7 August 2024 (UTC)

The trading and bartering guide image on the article needs to be updated; see specific details on the image page. The specific texture aren't that important as they're still recognizable, but we definitely need to add a note that Wind Burst can't be obtained from librarians. I don't know how to edit png files, so I need help. --MinecraftExp123^{(talk|contribs)} 09:18, 30 October 2024 (UTC)

As of 1.21.60.21, wandering traders do sell pale garden items in Bedrock. It appears to be the same price as in Java, however I couldn't add them in the table and in the history sections as:

• The regular/single offer mechanic in Bedrock confuses me.
• I'm unsure in which preview it was added (I don't think it has been mentioned in any changelog).

Thus this section could use of some help from someone knowledgeable. - Zamburger (talk) 11:19, 26 November 2024 (UTC)

Just did some testing, and at least with the trade rebalance option enabled, cartographer maps always point to the same structure in Bedrock Edition as well, at least for Jungle Explorer maps from a desert village cartographer. Was unsure whether or not I should remove/edit the info on the page.

2404:4401:9411:6700:7959:BD95:1454:92C7 00:47, 23 December 2024 (UTC)

As you can see in the image, I got two slots with Emerald > Iron armor, the slot no. 1 Coal > Emeralds is not in version 1.21.4. Nevertheless, I am unsure if I understood the table slotting correctly and ask for advice. Is the table wrong or something I misunderstood?

👁 Image

LevelRatings (talk) 01:08, 8 January 2025 (UTC)

You looked at the Bedrock Edition column that says 100% for coal. In Java Edition there's 40% probability for each trade 👁 Image
Miner(👁 Image
talk 👁 Image
contributions) 13:55, 8 January 2025 (UTC)

The wandering trader shows “saplings” listed as trades, does that mean all saplings? REXTHEREBEL228 (talk) 23:59, 19 January 2025 (UTC)

Probably all saplings and mangrove propagule 👁 Image
Miner(👁 Image
talk 👁 Image
contributions) 16:59, 15 February 2025 (UTC)

Thanks, sorry I could not get to this sooner! REXTHEREBEL228 (talk) 00:20, 16 February 2025 (UTC)

Wait, feedbacks go on talk pages? I thought it's only on Discord. I submitted this feedback. --MinecraftExp123^{(talk|contribs)} 03:06, 13 March 2025 (UTC)

It goes on talk pages so they can be tracked and marked as resolved Mudscape 👁 Image
talk 03:09, 13 March 2025 (UTC)

Oh. Sorry for putting something irrelevant on the talk page I guess. --MinecraftExp123^{(talk|contribs)} 03:10, 13 March 2025 (UTC)

The first factor is demand. An item that was sold out gets a price increase for all players when resupplied. If a player does not trade for a higher-priced item, the price is reduced the next time the villager resupplies. Demand is tracked per item, not per villager, so a villager can offer a higher-priced trade for a single item while other items are cheaper. Trades that have a price multiplier of 0 are not affected by demand. Arina (she/her) 07:54, 20 March 2025 (UTC)

If you're the same person who made the feedback in Ruined Portal, please refer to my message there on how to see old versions. The Minecraft Wiki documents the latest information for Minecraft.

However, for trading, I believe a page for trades before the implementation of Villager Trade Rebalance (when that releases) will be created.

Thanks for the feedback. — 3A |  T  C  09:09, 29 March 2025 (UTC)

Hi :)

The footnote for enchanted books from librarian trade reads:

The possible values are 5–19 emeralds for Lvl I, 8–32 for Lvl II, 11–45 for Lvl III, 14–58 for Lvl IV, and 17–71 for Lvl V. For treasure enchantments the price is doubled. The cost is capped at 64 emeralds. All values below this cap are equally probable.

A. It wasn't quite clear to me what the whole distribution looks like for level 5 trades and level 2 frost walker (the only ones affected by the capping as far as I can tell). On first reading, I thought it suggested that 17 to 64 are all equally likely for level 5, but I see now that that's not what it's suggesting, only that 17 to 63 are equally likely.

B. Treasure costs are even-valued only, but this wasn't clear from the reading.

---

On reading the code in 1.14 and 1.21.5 (sorry, I couldn't get the same link to generate the code for this version, but I got mine from following the fabric documentation here), the cost is uniformly sampled between 3L+2 and 13L+6, inclusive, where L is the enchantment level. This cost is then doubled for treasure enchantment (resulting in even-valued costs), and capped to 64.

So, for level 5 enchantments, each initial cost between 17 to 71 inclusive are equally likely with 1/55 chance. So the chance for 64 cost is 8/55 due to the cap. I.e. we expect 64-cost level 5 enchants 8 times more than any other costs.

I've included a test world download here with about 12k samples of level 5 enchanted book trades, counting how many times each cost appears. This results in about 1,900 64-cost ones, and the rest of the costs in the low 200's.

I think a similar exercise can be done for treasure enchantments if necessary. WaterGenie3 (talk) 10:33, 16 May 2025 (UTC)

I couldn't find more information about the demand in the wiki. But there's an old infographic here by MissLauralot: https://www.reddit.com/r/Minecraft/comments/d1a90d/ which is still mostly relevant, and I'll also be referencing the 1.21.5 code.

Demand vs Reputation

It may help to clarify somewhere that demands is tracked per trade listing per villager, as opposed to the gossip mechanics which is tracked per villager per player. In particular:

• Everyone else will also be subjected to the same effects when any player is making use of the trade listing.
• Multiple listing of the same item (e.g. enchanted books), regardless of whether they are on the same or different villagers, track their own demands separately.

Similar implications can be made for the gossiping mechanics:

• Only the player will see the effects of their own reputations on the villager (a common source of confusion in multiplayer after a player has already cured the villager).
• Player's reputation affects all trade listings of that villager for that player.

Demand

Each time a villager restock, the demand on each trade listing is:

1. Reduced by the maximum stock amount of that listing (maxUses), and
2. Increased by twice the number of trades made on that listing since the last restock (uses) [code].

This number of trades counts all of the instances made on that trade listing regardless of whether it was made individually or all in one operation.
There is no practical limit to the demand (java signed 32-bit integer limit) and it can go into the negatives.
When the demand is positive, the price on that trade listing will be increased by the base price x demand x price multiplier, rounded down. [code]
The final price is capped between 1 and the maximum stack size after adding the demand, reputation, and hero of the village components.

Some implications:

• Price does not necessarily increase or decrease with each restock since it depends on whether any increase puts the demand above 0 and high enough to get us to the next integer after multipling it by the price multiplier and rounding down, and whether any decrease puts the demand low enough to do the same but in the opposite direction.
• Not making use of a trade listing will also build up its demand more negative with each restock, so proportionately many trades are needed until we can drive it above 0, effectively building up a stockpile.
• 2 restock cancels out 1 max-out exactly, so making 1 max-out per in-game day keeps that trade's demand stable.

The formula for the reputation and hero of the village are already described in their respective pages.

So overall, the price is given by:

Clamp(
 base price
 + max[0, floor(base price × demand × price multiplier)] # Listing-based demand
 + ( -floor( ((5 × major pos) + minor pos + trade - minor neg - (5 × major neg)) × price multiplier) ) # Player-based reputation
 + ( -max[1, floor(base price × (0.0625 × (HotV level - 1) + 0.3))] ), # Player-based HotV
 1,
 maximum stack size
)

I think the demand mechanics is more appropriate in the villager supply/demand section so this page's sale price section can focus on the price calculation?

WaterGenie3 (talk) 01:31, 12 July 2025 (UTC)

Really? That sounds interesting. I couldn't find any related bugs, so either you are wrong, or there is an issue with our table. There is a pretty good chance of either of those being true. --Simanelix ^(T|C) 20:36, 7 June 2025 (UTC)

It was most likely a result of testing that did not have a large enough sample size. For example, when I tested the probability of an enchantment on bows, I tested like 3 or 4 (cant remember) full double chests of bows.
👁 Image
amethyst_hhh^{(talk)(weekly survey)} 21:06, 7 June 2025 (UTC)

Did you read my other comment about Template:TradeLine? It seems that it does not even support the extra data. --Simanelix ^(T|C) 21:52, 7 June 2025 (UTC)

Ah, no I didnt, I already had the page open before it was posted
👁 Image
amethyst_hhh^{(talk)(weekly survey)} 02:45, 8 June 2025 (UTC)

I tested it in-game and it is true. I'm not sure what the exact chances for trades are, but after testing it with a few villagers, the Journeyman trades seem to be equally distributed, and the Expert trades seem to have a higher chance of getting 2 banners than 1 banner and one item frame. It's probably just the same as in Java Edition but I can't verify that. MinecraftBedrockPlayer7 (talk) (contribs) 👁 Image
20:53, 7 June 2025 (UTC)

The Template:TradeLine does not seem to have parameters for Bedrock Edition's chances. I guess because Bedrock Edition uses slots and Java Edition does not and it is assumed that trade slots guarantee trades being there. --Simanelix ^(T|C) 21:04, 7 June 2025 (UTC)

The articles about trading, a mechanic exclusive to villagers and wandering traders, would not include information about piglins bartering a dried ghast. The bartering page does include that information though, please look there. - Harristic / Talk 👁 Image
16:58, 11 July 2025 (UTC)

Trading with piglins is something different than what this page is about; that's called bartering. MinecraftBedrockPlayer7 (talk) (contribs) 👁 Image
22:07, 27 July 2025 (UTC)

I've done about 30 attempts to get a fishing rod enchanted with something "good". All I have ever seen is Unbreaking I or Unbreaking II.

The note says that the algorithm chooses a "random enchantment" of levels 5-19. It also specifies no "treasure enchantments".

So, I SHOULD be able to get a level 3 enchantment of any of the three, Lure, Luck of the Sea, or Unbreaking. But all I am able to get is Unbreaking at levels 1 or 2.

Can someone confirm in the code if the note is correct or not? SirDaddicus (talk) 16:54, 31 July 2025 (UTC)

I’m not sure if it’s possible to get Unbreaking III at level 5–19. Either that, or you’re just unlucky.

“Level 5–19” is the same as applying something that requires 5–19 levels in an enchanting table, so you can use that as a reference. --MinecraftExp123^{(talk|contribs)} 08:11, 1 August 2025 (UTC)

My problem isn't so much the levels (although it SHOULD allow level III, given that librarians can and the text here on the wiki says so). My problem is that I've only gotten two enchantments (unbreaking I and II), No level III's, and no lure or luck of the sea.

They may have changed it as part of that nerf they almost implemented (trade "rebalance"). I need someone who can access the code to determine if that's true or not.

I'm up to 37 attempts, with only unbreaking and only levels 1 & 2. So unless someone shows that I'm wrong, I'm going to update the note to tell the truth. SirDaddicus (talk) 15:32, 2 August 2025 (UTC)

I've checked the code and they do chose a random enchantment of levels 5-19 just like all other trades that give enchanted items. I've then checked in game and I've got an Unbreaking III fishing rod on second try: https://i.imgur.com/MMK8pTA.png. Fisherman trades are also unaffected by the villager trade rebalance experiment.--Capopanzo (talk | contribs) 16:20, 2 August 2025 (UTC)

Java or Bedrock? I'm on Java. SirDaddicus (talk) 17:05, 13 August 2025 (UTC)

I'm also on Java.--Capopanzo (talk | contribs) 17:38, 13 August 2025 (UTC)

Thank you!

For the test and for confirming Java.

Now we need to know if unbreaking is the only option available Anybody else care to test? SirDaddicus (talk) 21:38, 13 August 2025 (UTC)

Unbreaking is not the only enchantment available; as you can see in the screenshot, you can also get something else (Lure in my case).--Capopanzo (talk | contribs) 21:46, 13 August 2025 (UTC)

Thanks. I can't see your screenshot, but I trust your judgment.

Must have been just rotten luck. I gave up after 35 tries. SirDaddicus (talk) 11:40, 10 September 2025 (UTC)

Why was the history table split up into different sections? 👁 Image
NmF (talk) 21:59, 3 August 2025 (UTC)

Because it makes the table much easier to navigate and read - especially on mobile. See also Forum:Compact history table. | violine1101 ^(talk) 22:03, 3 August 2025 (UTC)

The entry says 16, but I just locked out a cartographer with only 12 glass pane trades. I hadn't traded with this villager in ages. Can someone please verify this? Thanks! HolyT (talk) 05:17, 6 August 2025 (UTC)

It's indeed 12, this was changed in 1.21.5.--Capopanzo (talk | contribs) 21:24, 6 August 2025 (UTC)

Can someone check the code on how the price of enchanted iron pickaxes of wandering traders works? Is it random within the range, or does it depend on the enchantment, and if so, how? --MinecraftExp123^{(talk|contribs)} 13:32, 24 September 2025 (UTC)

In java code, the enchanted tools use the same function:

1. Each trade offer has a base price
   • This is specific to the trade offer, not the item (e.g. 1 for wandering trader's enchanted iron pick, but 3 for toolsmith's enchantd iron pick).
2. Each enchanted tool gets a random enchantment level uniformly distributed between 5 and 19, inclusive.
   • This is the "base enchantment level" B in this formula, the enchantments are generated in the same way described there.
3. The final price is the base price + the base enchantment level (e.g. 6-20 enchanted iron pick from wandering trader and 8-22 for toolsmith's).

---

There's a lot of randomness in how the enchantments are generated from the base enchantment level, so it is possible that we still get poor enchantments from a trade with higher enchantment level.

For example, according to this table, it's possible to just get unbreaking 3 from iron tool with base enchantment 19, and according this other table, it's possible to get efficiency 3 and unbreaking 3 from base enchantment 18, even though the 19 one will cost 1 higher emerald.

But in general, the more expensive, the more likely it is to get better enchantments.
For example, there's 15.23% chance of getting that eff3 unb3 from base enchantment 18, but it's 18.05% from base enchantment 19.

WaterGenie3 (talk) 09:21, 1 October 2025 (UTC)

I think that could be written as a footnote, with a reference to the note at all the relevant trades on the page. I'll do it soon when I have time. --MinecraftExp123^{(talk|contribs)} 09:26, 1 October 2025 (UTC)

I realise my wording is not entirely accurate:

When I just said "the more expensive, the more likely it is to get better enchantments", it's inaccurate in that the enchantment quality is proportional to only the random 5 to 19 part of the price, and not the whole price.

I.e. if we accrue negative gossips on a villager and then obtain an enchanted trade by rerolling/leveling up the villager, the trade will be even more expensive, but those price increases will not make the enchantment quality any higher.

Likewise, if we accrue positive gossips before hand, the trade will be even cheaper, but the price decreases will not make the enchantment quality any lower. WaterGenie3 (talk) 06:50, 2 October 2025 (UTC)

Okay, I understand. --MinecraftExp123^{(talk|contribs)} 06:51, 2 October 2025 (UTC)

@WaterGenie3 Finally, 👁 Image
 Done. I added footnotes to all the relevant trades. I guess I just procrastinated a lot. --MinecraftExp123^{(talk|contribs)} 03:42, 26 October 2025 (UTC)

Ohhh that's a nice way and accurate way of putting it, ty! WaterGenie3 (talk) 04:13, 26 October 2025 (UTC)

@WaterGenie3 Do the enchanted armor and sword trades use the same function too? --MinecraftExp123^{(talk|contribs)} 09:38, 1 October 2025 (UTC)

Yes, going in order as seen in the page, these all use the same function:

• Armorer's enchanted diamond gears
  • E.g. leggings has base price 14, boots 8, helmet 8, chestplate 16, and adding 5 to 19 to these should line up with their prices listed on the page.
• Fisherman's enchanted fishing rod
• Fletcher's enchanted bow/crossbow
• Toolsmith's enchanted iron/diamond tools
• Weaponsmith's enchanted weapons
• Wandering trader's enchanted iron pick

That's all the enchanted trades I could find. The only exception is librarian's enchanted book where the price is based on the enchantment's level instead.

This is still all in java though, I don't know enough about bedrock to check things there. WaterGenie3 (talk) 13:58, 1 October 2025 (UTC)

Villager XP is also missing.--Capopanzo (talk | contribs) 19:14, 30 October 2025 (UTC)

👁 Image
 Fixed | violine1101 ^(talk) 21:56, 30 October 2025 (UTC)

Price multiplier still appears to be missing. --MinecraftExp123^{(talk|contribs)} 23:11, 30 October 2025 (UTC)

@Violine1101 Can you fix it? --MinecraftExp123^{(talk|contribs)} 02:28, 31 October 2025 (UTC)

The order the items appear in the "Villager wants" column is also significant due to the pricing only affecting the first item (appears on the left in-game), at least in java.

Cases:

• Fletcher's novice level flint trade for 1 emerald and 10 gravel should have emerald as the first item.
  E.g. the master level tipped arrow trade for 2 emeralds and 5 arrows is already in the correct order.
  Curing a fletcher with these trades will yield cheaper emerald (still clamped at 1) + same 10 gravel for the former (as opposed to cheaper gravel + same emerald), and cheaper emerald + same 5 arrows for the latter.
• Fisherman's novice level cooked cod trade should have emerald + 6 raw cod.
  The apprentice level cooked salmon trade should have emerald + 6 raw salmon

WaterGenie3 (talk) 03:53, 31 October 2025 (UTC)

Finally, 👁 Image
 Done. I added the multiplier to the Lua module Module:Get trade info. ‑‑MinecraftExp123^{(talk|contribs)} 02:29, 13 December 2025 (UTC)

I am not experiencing this issue on either the mobile or desktop version of this page. BDJP ^(t|c) 23:11, 30 October 2025 (UTC)

This was probably an artifact of when I was tweaking things, it should be fixed now | violine1101 ^(talk) 00:25, 31 October 2025 (UTC)

Odds are supposed to be within each tier, but the table seems to incorrectly think the entire list of trades for each profession is a single tier. 👁 align=top
Sightnado ( talk / contribs ) 02:10, 5 November 2025 (UTC)

Yeah, it's a problem with Module:Get trade info, as it's supposed to transclude the tables from the individual profession pages, e.g. Armorer. I pinged Violine1101 at Template talk:Profession trades as they are the one that created the transcluding template. --MinecraftExp123^{(talk|contribs)} 02:21, 5 November 2025 (UTC)

Fixed by reverting some recent changes in Module:TradeTable.--Capopanzo (talk | contribs) 03:32, 5 November 2025 (UTC)

I was recently resetting trades with some villagers and noticed that the trade tables in the wiki are incorrect for bedrock edition in many places.

General formula for calculating the probability of each item being chosen for a given slot is 1 - ( ( m - 1 ) / m ) ^ n where m is the number of items to be selected from and n is the number of distinct slots to fill.

Based on an earlier topic in here, it seems like this is connected to certain modules that may require significant rework and updates across all villager types to get this corrected. I'm happy to help, but I'm not great with wiki coding.

Below is the correct set of trade tables for the Cartographer, as an example.

Editing to add that I have the wrong formula because I didn't realize there was a hard rule about duplicates not being allowed. I thought I saw two brown banners in one of the refreshed trades I looked at on a villager, but I must have been moving too fast.

Novice - wiki is correct

Trade table has a single group, select 2 of 2 options

wants	gives
paper	emerald
emerald	empty map

Apprentice - wiki is

Trade table has a single group, select 2 of 8 options - It should be 23.44% for each one.

wants	gives
glass pane	emerald
emerald+compass	explorer map - village_taiga
emerald+compass	explorer map - swamp_hut
emerald+compass	explorer map - village_snowy
emerald+compass	explorer map - village_savanna
emerald+compass	explorer map - village_plains
emerald+compass	explorer map - jungle_temple
emerald+compass	explorer map - village_desert

Journeyman - wiki is

Trade table has a single group, select 2 of 3 options - It should be 55.56% for each one.

wants	gives
compass	emerald
emerald+compass	exploration map - monument
emerald+compass	exploration map - trial_chambers

Expert - wiki is

Trade table has a single group, select 2 of 16 options - It should be 12.11% for each one.

Interestingly, id 7 for the banner is missing from the trade table. I am going to check if a bug report exists for this.

wants	gives
emerald	item frame
emerald	banner - id 4
emerald	banner - id 15
emerald	banner - id 1
emerald	banner - id 2
emerald	banner - id 10
emerald	banner - id 5
emerald	banner - id 6
emerald	banner - id 11
emerald	banner - id 14
emerald	banner - id 3
emerald	banner - id 13
emerald	banner - id 12
emerald	banner - id 9
emerald	banner - id 8
emerald	banner

Master - wiki is correct

Trade table has a single group, select 2 of 2 options

wants	gives
emerald	globe banner pattern
emerald+compass	exploration map - mansion

Lordicarus (talk) 22:39, 18 December 2025 (UTC)

The bedrock probabilities are tied to the slots in which the item can appear (see Cartographer#Trades, they're hidden on this page for... technical reasons). Is this no longer the case? In which version did you test this? | violine1101 ^(talk) 22:52, 18 December 2025 (UTC)

Cartographers are indeed currently incorrect as we don't support biome-specific trades, although the probabilities posted by OP also seem wrong. Different biome variants sell different amounts of explorer maps and banners, which affect the probability for glass panes in the apprentice group and item frames in the expert group. Apprentice, journeyman and expert trades are also not tied to specific slots, similarly to Java. The Chinese wiki has the correct probability values: zh:制图师#交易.--Capopanzo (talk | contribs) 23:00, 18 December 2025 (UTC)

I've also just checked the newly added data pack definitions for trades in Java Edition 26.1 Snapshot 1, and it seems like it also does some stuff affecting the probabilities which we don't account for (e.g. allowing duplicates in trade sets), which we don't account for. So Java Edition probabilities are likely incorrect as well. | violine1101 ^(talk) 23:02, 18 December 2025 (UTC)

Nevermind, duplicates are never allowed in vanilla, it's just an extra datapack feature. But the biome thing also affects Java. | violine1101 ^(talk) 23:07, 18 December 2025 (UTC)

My installation is vanilla. Nothing is installed that would be affecting any of my trades. Lordicarus (talk) 23:14, 18 December 2025 (UTC)

Just ack that I was wrong about the duplicates here and wasn't paying attention to that aspect. Lordicarus (talk) 00:00, 19 December 2025 (UTC)

Those probability values are not mathematically correct on the Chinese wiki.

When you have 3 slots to fill and 2 potential items, where duplicates are allowed, you don't end up with a 67% chance for each item. You have a 55.56% chance of getting each item. That is the formula I included above. You can see an article that explains the math behind this here: https://www.storyofmathematics.com/probability-with-replacement/ Lordicarus (talk) 23:13, 18 December 2025 (UTC)

Duplicates are not allowed, if an item is selected as the first trade it cannot be picked for the other trade. Also the chances you provided for apprentice and expert trades are wrong because they are assuming a cartographer can sell any biome map and any banner color, which is not the case; e.g. a plains cartographer can only sell savanna or taiga maps, while a desert cartographer can sell 3 different maps, affecting the chances for the glass panes trade as a result.--Capopanzo (talk | contribs) 23:36, 18 December 2025 (UTC)

Ah yes, I see the variant requirement. Thanks for that. Also, if duplicates are not allowed, then the percentages I provided are not correct, but the percentages on the wiki still need to be updated. Also, the item frame and glass panes do not seem to have restrictions based upon villager biome variant (according to the JSON) like the banners have.

https://imgsh.net/i/1e4b294b03

Thanks for steering me in the correct direction. Lordicarus (talk) 23:58, 18 December 2025 (UTC)

This is in 1.21.130. I have confirmed through actual villager trade resets with a res pack that allows me to see all trades for a particular villager without having to manually unlock them. For a cartographer I just worked on, the journeyman level was not giving me a trail map.

When I looked at the actual trades in the data files (behavior_pack\trading\economy_trades\cartographer_trades.json) I can see that the way the trades are grouped, there are not any slots in the way the wiki outlines it.

Here is the code for the journeyman level trades for cartographer. You can see in the code that there aren't individual slots with the way they are doing it now. Lordicarus (talk) 23:02, 18 December 2025 (UTC)

https://minecraft.wiki/w/Talk:Trading#c-Lordicarus-20251218230200-Violine1101-20251218225200 Lordicarus (talk) 23:02, 18 December 2025 (UTC)

Sorry pasted the wrong link. https://pastebin.com/yv5LCjhd Lordicarus (talk) 23:03, 18 December 2025 (UTC)

--Lordicarus (talk) 00:06, 19 December 2025 (UTC)

Enchanted books are the most important trade of the librarian, and arguably of any villager. Information about enchantment levels and emerald costs should have its own section, or at least its own table, rather than being relegated to a mere note. The way the information is formatted now makes it hard to find, and harder to parse. OriginalAwawe (talk) 14:08, 19 February 2026 (UTC)

Hello,

The sales prices formula given at the end of the Mechanics section is incorrect. I think the last should be instead. For instance, let the Hero of the Village level be . Then the formula condenses into

, which would evaluate to 1 in default cases.

If I were autoconfirmed, I would have fixed that, since a not-so-clearly incorrect formula is better than a clearly incorrect one. ॥ ☼ SunnyMoon ☽ ॥ (talk) 11:51, 24 February 2026 (UTC)

Can you write out what the correct formula would be? I don't really understand what you meant by "the last ". ‑‑MinecraftExp123^{(talk|contribs)} 12:24, 24 February 2026 (UTC)

Sure! I meant

.

I meant that the last in the entire equation should be . ॥ ☼ SunnyMoon ☽ ॥ (talk) 12:29, 24 February 2026 (UTC)

Oh, that does seem logical for HotV level zero, but for levels 1 and greater, that would make evaluate to 1, since would be greater than 1 for most prices. I'll double check and see how to fix it. ‑‑MinecraftExp123^{(talk|contribs)} 12:34, 24 February 2026 (UTC)

@SunnyMoon I think I fixed it. Firstly, the entire Hero of the Village term is wrong, because it originally gave the discounted price, not the discount itself, as it used to decrease as the HotV level went up. Now I fixed it so that it increases. Then, I fixed what you pointed out by multiplying the discount by , which removes the discount because it becomes 0 when the HotV level is 0. ‑‑MinecraftExp123^{(talk|contribs)} 12:57, 24 February 2026 (UTC)

Great! This looks better. :) ॥ ☼ SunnyMoon ☽ ॥ (talk) 13:01, 24 February 2026 (UTC)

Everything seems correct on the chart. What do you think is wrong? 👁 Image
Redstone Engineer ^{(Talk) (Survey)} 17:30, 1 April 2026 (UTC)

Cartographer, librarian and wandering traders are outdated. --Capopanzo (talk | contribs) 18:07, 1 April 2026 (UTC)

i had 15 villagers that i turned into farmers, and all of them were selling pumpkins on their second lvl up, wiki article says its 67%, but im not really sure on that Protag (talk) 18:09, 7 April 2026 (UTC)

It's correct, in Java apprentice farmers pick 2 trades out of 3 possible choices (buy pumpkins, sell pumpkin pies, sell apples), and any trade within the same pool has the same chance of being picked. You simply got lucky. --Capopanzo (talk | contribs) 18:48, 7 April 2026 (UTC)

It's in the correct slot. From Cartographer#Biome-dependent probabilities:

Glass panes and item frames can be sold by cartographers from any biome, but the probabilities of these trades depend on the biome the villager is from. This is because the amount of explorer maps or banners they can sell varies based on the villager's biome; this changes the total amount of trades they can pick from, affecting the chance of each individual trade at apprentice or expert level. Capopanzo (talk | contribs) 14:23, 15 April 2026 (UTC)

When checking villager trades, the table shows the default values for the materials the Villagers ask for, but if one wants to check how much trades would cost with modifiers, for example if the villager was cured from a zombie, then one would have to go to a separate page, grab the modifiers, and manually calculate the values for each item.

Just like block pages have Template:Breaking row, a table that shows the breaking time for blocks with sliders for the Efficiency enchantment and Haste effect levels to see how those affect the time, Villager trades should become a template with options to select if a villager has been cured, if the player has one of 5 levels of Hero of the Village, and even the gossip level such as the negative effects of hitting/killing a villager or the positive from trading with them repeatedly. A template would also allow separation of trade differences between Java and Bedrock, one could choose which to view so both options aren't present on the same chart. 👁 Image
404_11 (talk) 15:59, 9 January 2026 (UTC)

👁 Image
Really, Really, Strong Support I’ve also wondered about this. I also think a calculator for trade price would also help 👁 Image
Max Dirt Eater Alt 👁 Image
01:24, 21 June 2026 (UTC)

👁 Image
 Strong support as this is just very practical and useful. Maybe this can be made into a calculator (as a subpage of Calculators). ‑‑MinecraftExp123^{(talk|contribs)} 02:42, 21 June 2026 (UTC)

Yeah, that’s what I meant by calculators. Also, how should we find the modifiers? Check the code? 👁 Image
Max Dirt Eater Alt 👁 Image
02:48, 21 June 2026 (UTC)

This page already has many details, though if any info is missing, sure. It's available at https://mcsrc.dev ‑‑MinecraftExp123^{(talk|contribs)} 02:49, 21 June 2026 (UTC)

Cool! Also, how do you make calculators. I am not quite sure and don’t plan to make any (cause my JavaScript sucks) but I just want to know 👁 Image
Max Dirt Eater Alt 👁 Image
02:54, 21 June 2026 (UTC)

You can use the source code of the armor protection calculator as a reference (which I made a while ago). ‑‑MinecraftExp123^{(talk|contribs)} 02:56, 21 June 2026 (UTC)

Okay thanks! 👁 Image
Max Dirt Eater Alt 👁 Image
03:32, 21 June 2026 (UTC)

Retrieved from "https://minecraft.wiki/w/Talk:Trading?oldid=3638841"